John Spray c84656a53e pageserver: implement auto-splitting (#7681) ...

## Problem

Currently tenants are only split into multiple shards if a human being
calls the API to do it.

Issue: #7388 

## Summary of changes

- Add a pageserver API for returning the top tenants by size
- Add a step to the controller's background loop where if there is no
reconciliation or optimization to be done, it looks for things to split.
- Add a test that runs pgbench on many tenants concurrently, and checks
that splitting happens as expected as tenants grow, without interrupting
the client I/O.

This PR is quite basic: there is a tasklist in
https://github.com/neondatabase/neon/issues/7388 for further work. This
PR is meant to be safe (off by default), and sufficient to enable our
staging environment to run lots of sharded tenants without a human
having to set them up.

2024-05-17 16:01:24 +00:00

..

data/extension_test/5670669815

Feat/postgres 16 (#4761)

2023-09-12 15:11:32 +02:00

test_ancestor_branch.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_attach_tenant_config.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_auth.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_backpressure.py

Revert "pageserver: use a single tokio runtime (#6555)" (#7246)

2024-03-26 15:24:18 +01:00

test_bad_connection.py

Add retry loops and bump test timeout in test_pageserver_connection_stress (#7281)

2024-05-01 21:36:50 -07:00

test_basebackup_error.py

Rename "Postgres nodes" in control_plane to endpoints.

2023-04-13 14:34:29 +03:00

test_branch_and_gc.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_branch_behind.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_branching.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_broken_timeline.py

pageserver: remove Option<> around remote storage, clean up metadata file refs (#7752)

2024-05-15 12:05:24 +00:00

test_build_info_metric.py

feat: add build_tag env support for set_build_info_metric (#5576)

2023-10-27 10:47:11 +01:00

test_change_pageserver.py

tests/neon_local: rename "attachment service" -> "storage controller" (#7087)

2024-03-12 11:36:27 +00:00

test_clog_truncate.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_close_fds.py

tests: enable multiple pageservers in neon_local and neon_fixture (#5231)

2023-09-08 16:19:57 +01:00

test_compaction.py

Add test_uploads_and_deletions test (#7758)

2024-05-15 15:32:47 +02:00

test_compatibility.py

chore(test): add version check for forward compat test (#7685)

2024-05-14 10:36:48 -04:00

test_compute_catalog.py

compute_ctl: catalog API endpoints (#7575)

2024-05-16 12:04:16 +02:00

test_config.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_crafted_wal_end.py

Allow bad state (not active) pageserver error/warns in walcraft test.

2024-05-06 06:45:27 +03:00

test_createdropdb.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_createuser.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_ddl_forwarding.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_disk_usage_eviction.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_download_extensions.py

Use test specific directory in test_remote_extensions (#5938)

2023-11-27 18:57:58 +00:00

test_duplicate_layers.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_explain_with_lfc_stats.py

new test for LFC stats in explain (#6968)

2024-03-01 14:33:08 +00:00

test_fsm_truncate.py

Rename "Postgres nodes" in control_plane to endpoints.

2023-04-13 14:34:29 +03:00

test_fullbackup.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_gc_aggressive.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_hot_standby.py

Add test to demonstrate the problem with protocol version 1 (#7377)

2024-04-25 20:45:37 +03:00

test_import.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_large_schema.py

test_runner: replace all .format() with f-strings (#7194)

2024-04-02 14:32:14 +01:00

test_layer_bloating.py

test_runner: replace all .format() with f-strings (#7194)

2024-04-02 14:32:14 +01:00

test_layer_eviction.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_layer_writers_fail.py

Move tenant & timeline dir method to NeonPageserver and use them everywhere (#5262)

2023-09-15 11:17:18 +01:00

test_layers_from_future.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_lfc_resize.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_lfc_working_set_approximation.py

Testcase for neon extension function approximate_working_set_size() (#6980)

2024-03-01 13:29:08 +01:00

test_local_file_cache.py

pageserver: coalesce read paths (#7477)

2024-04-25 13:29:17 +01:00

test_logging.py

tests: add optional cursor to log_contains + fix truthiness issues in callers (#6960)

2024-03-01 10:45:39 +01:00

test_logical_replication.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_lsn_mapping.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_migrations.py

Revoke REPLICATION (#7052)

2024-03-08 22:24:30 +00:00

test_multixact.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_neon_cli.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_neon_extension.py

spec: allow neon extension auto-upgrade + softfail upgrade (#7231)

2024-03-28 17:22:35 +00:00

test_neon_local_cli.py

fix(test suite): some tests leak child processes (#6497)

2024-01-26 18:23:53 +00:00

test_neon_superuser.py

fix(test): drop subscription when test completes (#6975)

2024-03-06 15:52:24 +00:00

test_next_xid.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_normal_work.py

Rename "Postgres nodes" in control_plane to endpoints.

2023-04-13 14:34:29 +03:00

test_old_request_lsn.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_ondemand_download.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_ondemand_slru_download.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_pageserver_api.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_pageserver_catchup.py

Rename "Postgres nodes" in control_plane to endpoints.

2023-04-13 14:34:29 +03:00

test_pageserver_generations.py

pageserver: fixes for layer path changes (#7786)

2024-05-17 13:24:03 +01:00

test_pageserver_getpage_throttle.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_pageserver_layer_rolling.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_pageserver_metric_collection.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_pageserver_reconnect.py

Implement lockless update of pageserver_connstring GUC in shared memory (#6314)

2024-01-23 07:55:05 +02:00

test_pageserver_restart.py

pageserver: label tenant_slots metric by slot type (#7603)

2024-05-06 14:07:15 +01:00

test_pageserver_restarts_under_workload.py

pageserver: improve the shutdown log error (#5792)

2023-11-07 16:57:26 +00:00

test_pageserver_secondary.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_parallel_copy.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_pg_regress.py

pageserver: coalesce read paths (#7477)

2024-04-25 13:29:17 +01:00

test_pg_waldump.py

pg_waldump segment on safekeeper in test_pg_waldump.

2024-05-06 07:18:38 +03:00

test_physical_replication.py

Track size of FSM fork while applying records at replica (#5901)

2023-12-05 18:49:24 +02:00

test_pitr_gc.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_postgres_version.py

Add Postgres commit sha to Postgres version (#4603)

2024-05-07 15:18:17 +00:00

test_proxy_allowed_ips.py

proxy: include client IP in ip deny message (#6854)

2024-02-21 18:24:59 +01:00

test_proxy_metric_collection.py

refactor(test_consumption_metrics): split for pageserver and proxy (#5324)

2023-09-16 18:05:35 +03:00

test_proxy_websockets.py

proxy: swap tungstenite for a simpler impl (#7353)

2024-05-16 13:05:50 +02:00

test_proxy.py

remove http timeout (#7291)

2024-04-03 11:23:26 +01:00

test_read_trace.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_read_validation.py

Refactor how the request LSNs are tracked in compute (#7377)

2024-04-25 20:45:37 +03:00

test_readonly_node.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_recovery.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_remote_storage.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_replication_start.py

fixup(#7204 / postgres): revert IsPrimaryAlive checks (#7209)

2024-03-23 01:01:51 +00:00

test_s3_restore.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_s3_scrubber.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_setup.py

python: more linting (#4734)

2023-07-18 12:56:40 +03:00

test_sharding.py

pageserver: implement auto-splitting (#7681)

2024-05-17 16:01:24 +00:00

test_sni_router.py

tests: split neon_fixtures.py (#4871)

2023-08-03 17:20:24 +03:00

test_storage_controller.py

Fix test_lock_time_tracing flakiness (#7712)

2024-05-13 13:18:14 +01:00

test_subxacts.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_tenant_conf.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_tenant_delete.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_tenant_detach.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_tenant_relocation.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_tenant_size.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_tenant_tasks.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_tenants_with_remote_storage.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_tenants.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_threshold_based_eviction.py

tests: add optional cursor to log_contains + fix truthiness issues in callers (#6960)

2024-03-01 10:45:39 +01:00

test_timeline_delete.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_timeline_detach_ancestor.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_timeline_size.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_truncate.py

python: more linting (#4734)

2023-07-18 12:56:40 +03:00

test_twophase.py

tests: Remove "postgres is running on ... branch" messages

2024-02-11 01:34:31 +02:00

test_unlogged.py

Rename "Postgres nodes" in control_plane to endpoints.

2023-04-13 14:34:29 +03:00

test_vm_bits.py

Ignore page header when comparing VM pages in test_vm_bits.py (#7499)

2024-05-08 20:58:35 +03:00

test_wal_acceptor_async.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_wal_acceptor.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_wal_receiver.py

test_suite: patch, don't replace, the tenant_config field, where appropriate (#7771)

2024-05-17 12:24:02 +02:00

test_wal_restore.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00

test_walredo_not_left_behind_on_detach.py

python: allow using allowed_errors.py (#7719)

2024-05-13 15:16:23 +03:00